Abstract: Briefly, embodiments of methods and/or systems of training multiclass convolutional neural networks (CNNs) are disclosed. For one embodiment, as an example, an auxiliary CNN may be utilized to form an ensemble with the collection as a linear combination. The linear combination may be based, at least in part, on boost prediction error encountered during the training process.

Abstract: A search engine for finding objects that correspond to a search request, including an input module for receiving a keyword query from a user, and a search module being configured to map the keyword query to the identifiers of objects that semantically match the keyword or the plurality of keywords contained in the keyword query, and to generate a search result that contains a listing of matching object identifiers, is characterized in that the search module is further configured to generate the search result by considering network layer information about the user within the process of mapping the keyword query to identifiers of matching objects, wherein the network layer information include sophisticated information the search module receives from a dedicated entity.

Abstract: Methods for categorizing news are presented. One method groups articles into clusters that share a common topic. A first category is identified for each article that indicates if the article is news or not. Further, the method includes an operation for determining use data for each article that has information about people that have accessed or referenced the article. Additionally, the method includes an operation for combining the use data and the first category for all the articles in each cluster to determine the geographical scope of interest for the cluster. The use data and the first category are combined for all the articles in each cluster to determine a second category for each article that indicates if the article is general news, topical news, or not news. The articles are presented to the user based on the geographical scope of interest, the second category, and the attributes of the user.

Abstract: According to one aspect of an embodiment a learning device includes a generating unit that generates a new second learner by using a part of a first learner in which deep learning has been performed on the relationship held by a combination of first content and second content that has a type different from that of the first content. The learning device includes a learning unit that allows the second learner generated by the generating unit to perform deep learning on the relationship held by a combination of the first content and third content that has a type different from that of the second content.

Abstract: Disclosed herein are aspects associated with contextual, or related, media enrichment presentation item of a media object served via the internet. A request to annotate a media object in connection with the media object's presentation is received, and a media object identifier and a profile identifier are obtained. The media object's information is retrieved using the media object identifier, and a profile is retrieved using the profile identifier. A response including one or more references to one or more media enrichment presentation items is transmitted, each reference to a media enrichment presentation item comprising information for use in retrieving the media enrichment presentation item for presentation in connection with presentation of the media object.

Abstract: Systems and methods to define a social network for a user using data implied by communications of the user with others. In one embodiment, a method includes scanning messages to identify persons addressed in the messages; computing scores of the persons to determine a relevancy of each person; receiving, by a first computing device, a request from a computer program running on a second computing device; in response to the request from the computer program, providing access to friend data for the persons, the computer program to provide services based on the friend data; and sending the friend data to the second computing device.

Abstract: Disclosed are systems and methods for improving interactions with and between computers in content generating, searching, hosting and/or providing systems supported by or configured with personal computing devices, servers and/or platforms. The systems interact to identify and retrieve data within or across platforms, which can be used to improve the quality of data used in processing interactions between or among processors in such systems. The disclosed systems and methods provide systems and methods for automatically extracting and creating an animated Graphics Interchange Format (GIF) file from a media file. The disclosed systems and methods identify a number of GIF candidates from a video file, and based on analysis of each candidate's attributes, features and/or qualities, as well as determinations related to an optimal playback setting for the content of each GIF candidate, at least one GIF candidate is automatically provided to a user for rendering.

Abstract: Software running on servers at a website hosting a news service generates a first profile for a user of the news service. The first profile is based at least in part on implicit relevance feedback from the user on content presented by the news service. The software obtains a second profile for the user from a web-searching service. The software creates a score for a candidate item of content. The score is based on similarity of the candidate item to the first profile and similarity of the candidate item to the second profile. Similarity to the second profile measures at least similarity to a plurality of web-search queries and similarity to any titles of any search results resulting from each of the queries. The software then presents the item of content to the user in a content stream served by the news service, based on the score.

Abstract: A distribution device includes a distribution unit configured to distribute, to a user terminal, a content control program for controlling content displayed on the user terminal. The content control program causes the user terminal to execute a viewable area acquisition procedure of acquiring, as information on a viewable area, information on an area of the content displayed by a browser program, a reference position acquisition procedure of acquiring pieces of information on a plurality of reference positions defined on a display area of the content on which a video image is displayed, a determination procedure of determining whether the plurality of reference positions are located in the viewable area, and a playback start procedure of starting playback of the video image when it is determined that the plurality of reference positions are located in the viewable area.

Abstract: A search-enabled remote control device to facilitate viewers' ability to search through programming and/or content options and view results. In one embodiment, a viewer is able to use a traditional remote control interface to navigate through menus or selections and to display search results on screen. In another embodiment, a viewer's watching behavior and/or searching habits are tracked by the search-enabled remote control and used to develop a viewer profile, which may in turn be used to target advertisements, information and/or future programming options.

Abstract: An example system can include a server that includes or is associated with a keyword recommendation module. The keyword recommendation module can be configured to select keywords for a search engine for use in a search engine marketing campaign, wherein the search engine provides more traffic to Internet content of a publisher than other search engines, for the keywords. The module can also be configured to determine, per keyword, user engagement with the Internet content resulting from the traffic provided by the search engine for the keywords, according to one or more of time spent viewing the Internet content, page views of the Internet content, and dwell times. The module can also be configured to score, per keyword, the keywords according to the determined user engagement with the Internet content, and generate keyword recommendations according to the scoring of the keywords.

Abstract: Methods and structures are disclosed that allow a user navigation action (e.g., clicking a link in a web browser) to be intercepted for a determination of whether to provide the user an alternate page instead of a requested target page. A user may be redirected to an alternate page based on various criteria, including data related to content of the target page, whether the target page is unavailable, etc. In some embodiments, alternate page determination is performed by a computer system of a website different from a website of the target page. Thus, presenting an alternate page to a user may be based on a target page featuring an out-of-stock product, or one that is currently unavailable with certain features. The alternate page may include more useful content, and may include a current link to purchase the product, a link to purchase a similar product, or other features.

Abstract: Briefly, methods and/or systems of processing content entries are described. An example may comprise determining equivalent byte values of characters that form the content entries. The content entries may be transformed based, at least in part, on the equivalent byte values and compressed using, for example, delta compression.

Abstract: Techniques are provided for integrating rich media into a scalable, multi-user, web-based calendar system having multiple interconnected web-based interfaces. Rich media refers to digital media that exhibit either, or a combination, of dynamic or interactive properties, including video content, audio content, graphical content, hyperlinks, or browser-based applications to such content, that are associated with a particular calendar event. Rich media events are created and stored in a central repository, and are accessible for adding to users' own calendar events. Rich media calendar events include display templates that define display characteristics for the rich media calendar event. Display templates also provide rich media applications for rendering and displaying the rich media content in the event. Delivery schedules are employed to automatically control the display of rich media calendar events.

Abstract: Methods and system for generating companion content for rendering on a secondary device during presentation of content on a main device include detecting execution of a local application on the secondary device by a server application. The local application establishes a wireless connection through an application programming interface (API) with the secondary device and a device that interfaces with the main device and accesses metadata defining content being rendered on the main device. The local application transmits the metadata received from the device to the server application. Historical data is generated based on the metadata received from the secondary device. The metadata within the historical data is analyzed to identify surfing paths of one or more users over time. Companion content is generated based on the surfing paths of the one or more users and forwarded to the secondary device for rendering while the secondary device is connected to the device that interfaces with the main device.

Abstract: An output resource identifier, such as a universal resource locator (URL), may be programmatically generated using one or more recomposition rules and decomposed parts of a source, or input, URL. The decomposed parts may be programmatically generated using one or more decomposition rules. The input and output URLs may comprise one or more of web and/or native URLs.

Abstract: Disclosed are methods and apparatus for selecting items (e.g., Internet memes) to be presented to a user. In one embodiment, a user connected to a network is identified. For each of a plurality of items (e.g., memes), for each of one or more followers of the user, a probability that the corresponding one of the followers of the user will repost the one of the plurality of items is determined. A number of the plurality of items may be selected based, at least in part, on the probability that has been determined, for each of the plurality of items, for each of the one or more followers of the user, that the corresponding one of the followers of the user will repost the one of the plurality of items. The selected number of the plurality of items may then be presented to the user.

Abstract: A system can effectively compress instrumentation data related to a page view on a client-side application thread (such as a web browser thread) and communicate the compressed data over the Internet to a server (such as an analytics server). Leading up to the compression, the system can include data scanning, code building, code encoding, and code serialization. The compression can run on a background thread of the client-side application, so that it does not interfere with other processes run by the client-side application. For example, a web worker can implement the code compression described herein.

Abstract: As provided herein, a user of a client device may navigate to a webpage using a browser. A browser window, populated with a verification image and/or details about the webpage, is generated and presented to the user. The verification image and/or details about the webpage differentiate a browser window generated by the browser, from the webpage, from a browser window generated by a malicious user. The browser window comprises a login box into which credentials for logging into the user account may be entered. Responsive to the user entering correct credentials into the login box and selecting a submit option based upon recognition of the verification image, the browser window may be submitted to a server and the user may be presented with a window comprising access to the user account.